A guest blog from Prof Dr Marjanka Schmidt, a genetic breast cancer epidemiologist with a senior group leader position at the Netherlands Cancer Institute – Antoni van Leeuwenhoek hospital and a chair at the Leiden University Medical Center, The Netherlands. She is the principal investigator and coordinator of the B-CAST project, in which PHG Foundation is a partner.
We probably all know someone (who knows someone) that has breast cancer, or has survived breast cancer. Breast cancer is the top prevalent disease among women worldwide. We can do some things to reduce the risk of developing breast cancer, such as having a healthy weight, being active, drinking little alcohol. However, many breast cancers are caused by factors we have little control over: ageing, naturally occurring hormones, hereditary factors.
As with many other cancers, once breast cancer is diagnosed, there are good treatment options such as surgery with or without radiotherapy, and chemotherapy and/or endocrine treatment (in the form of pills or intravenous drip) that may lead to cure. Cure is most likely if a breast cancer is found at an early stage. Therefore, a good strategy is to make sure we find breast cancers early enough such that they can still be treated successfully. Moreover, the earlier we find them, the less aggressive treatment needs to be, with fewer side-effects and loss of health.
Research over the last years has identified several factors we can use to identify women at risk of breast cancer. These include hereditary factors based on family history or mutations in breast cancer genes, such as BRCA1/2 or PALB2, CHEK2, and ATM; and other factors such as high breast density, hormone exposure and lifestyle factors. Combining such factors in risk prediction models, we can try to predict which women are highly likely to develop breast cancer, and which women are at really low risk.
With help of EU funding, in the B-CAST project, we updated and validated comprehensive risk prediction models, such as the CanRisk model. This is a first step in the right direction.
Models such as CanRisk could be used to stratify women in national (mammography) screening programmes for more or less intensive screening. They could also help women who have been identified as at high risk of breast cancer and who then visit a clinical genetic centre for advice regarding risk reducing surgery.
Risk stratification can also benefit women at low risk, who may need less intensive screening, with the potential to reduce unnecessary biopsies and scans.
All this promising news. However – we can only find early signs of breast cancers if we are looking for them i.e. if the women affected are screened in time. But most national screening programs are for women above age 50 years. Therefore, we miss an important group of women. Young women, who are not yet eligible for national screening programmes. Young women, perhaps from small families with no first- or second-degree family members with breast cancer, who are not being referred to clinical genetics for screening of breast cancer genes. These are women that enter our hospitals with breast cancer, often in an aggressive form, and who are sometimes found to have a mutation in the high risk BRCA1 or other genes.
We need to close the gap between the screening and risk stratification in clinical genetics, and population screening in national programmes. Firstly, by identifying the women with high risk of breast cancer that are being missed in our screening programmes; secondly, by better understanding the causal factors for their disease; and thirdly, by tailoring not only our models and tools but also our screening programmes to allow the prediction, early detection, and classification of a breast cancer in these women.
The breast tumors we most urgently need to find are those that are aggressive and potentially fatal. Cost-effective tools that will help us to find breast cancer early enough to optimise treatment outcomes reduces the burden of breast cancer treatment for women and health services, and can save more lives. Research like the B-CAST programme is helping develop such tools.